Public Response to Federal Electronic Cigarette Regulations Analyzed Using Social Media Data Through Natural Language Processing: Topic Modeling Study.

BACKGROUND: e-Cigarette (electronic cigarette) use has been a public health issue in the United States. On June 23, 2022, the US Food and Drug Administration (FDA) issued marketing denial orders (MDOs) to Juul Labs Inc for all their products currently marketed in the United States. However, one day later, on June 24, 2022, a federal appeals court granted a temporary reprieve to Juul Labs that allowed it to keep its e-cigarettes on the market. As the conversation around Juul continues to evolve, it is crucial to gain insights into the sentiments and opinions expressed by individuals on social media. OBJECTIVE: This study aims to conduct a comprehensive analysis of tweets before and after the ban on Juul, aiming to shed light on public perceptions and sentiments surrounding this contentious topic and to better understand the life cycle of public health-related policy on social media. METHODS: Natural language processing (NLP) techniques were used, including state-of-the-art BERTopic topic modeling and sentiment analysis. A total of 6023 tweets and 22,288 replies or retweets were collected from Twitter (rebranded as X in 2023) between June 2022 and October 2022. The encoded topics were used in time-trend analysis to depict the boom-and-bust cycle. Content analyses of retweets were also performed to better understand public perceptions and sentiments about this contentious topic. RESULTS: The attention surrounding the FDA's ban on Juul lasted no longer than a week on Twitter. Not only the news (ie, tweets with a YouTube link that directs to the news site) related to the announcement itself, but the surrounding discussions (eg, potential consequences of this ban or block and concerns toward kids or youth health) diminished shortly after June 23, 2022, the date when the ban was officially announced. Although a short rebound was observed on July 4, 2022, which was contributed by the suspension on the following day, discussions dried out in 2 days. Out of the top 50 most retweeted tweets, we observed that, except for neutral (23/45, 51%) sentiment that broadcasted the announcement, posters responded more negatively (19/45, 42%) to the FDA's ban. CONCLUSIONS: We observed a short life cycle for this news announcement, with a preponderance of negative sentiment toward the FDA's ban on Juul. Policy makers could use tactics such as issuing ongoing updates and reminders about the ban, highlighting its impact on public health, and actively engaging with influential social media users who can help maintain the conversation.

Characterizing pituitary adenomas in clinical notes: Corpus construction and its application in LLMs.

Objective: Faced with the challenges of differential diagnosis caused by the complex clinical manifestations and high pathological heterogeneity of pituitary adenomas, this study aims to construct a high-quality annotated corpus to characterize pituitary adenomas in clinical notes containing rich diagnosis and treatment information. Methods: A dataset from a pituitary adenomas neurosurgery treatment center of a tertiary first-class hospital in China was retrospectively collected. A semi-automatic corpus construction framework was designed. A total of 2000 documents containing 9430 sentences and 524,232 words were annotated, and the text corpus of pituitary adenomas (TCPA) was constructed and analyzed. Its potential application in large language models (LLMs) was explored through fine-tuning and prompting experiments. Results: TCPA had 4782 medical entities and 28,998 tokens, achieving good quality with the inter-annotator agreement value of 0.862-0.986. The LLMs experiments showed that TCPA can be used to automatically identify clinical information from free texts, and introducing instances with clinical characteristics can effectively reduce the need for training data, thereby reducing labor costs. Conclusion: This study characterized pituitary adenomas in clinical notes, and the proposed method were able to serve as references for relevant research in medical natural language scenarios with highly specialized language structure and terminology.

Scalable incident detection via natural language processing and probabilistic language models.

Post marketing safety surveillance depends in part on the ability to detect concerning clinical events at scale. Spontaneous reporting might be an effective component of safety surveillance, but it requires awareness and understanding among healthcare professionals to achieve its potential. Reliance on readily available structured data such as diagnostic codes risks under-coding and imprecision. Clinical textual data might bridge these gaps, and natural language processing (NLP) has been shown to aid in scalable phenotyping across healthcare records in multiple clinical domains. In this study, we developed and validated a novel incident phenotyping approach using unstructured clinical textual data agnostic to Electronic Health Record (EHR) and note type. It's based on a published, validated approach (PheRe) used to ascertain social determinants of health and suicidality across entire healthcare records. To demonstrate generalizability, we validated this approach on two separate phenotypes that share common challenges with respect to accurate ascertainment: (1) suicide attempt; (2) sleep-related behaviors. With samples of 89,428 records and 35,863 records for suicide attempt and sleep-related behaviors, respectively, we conducted silver standard (diagnostic coding) and gold standard (manual chart review) validation. We showed Area Under the Precision-Recall Curve of ~ 0.77 (95% CI 0.75-0.78) for suicide attempt and AUPR ~ 0.31 (95% CI 0.28-0.34) for sleep-related behaviors. We also evaluated performance by coded race and demonstrated differences in performance by race differed across phenotypes. Scalable phenotyping models, like most healthcare AI, require algorithmovigilance and debiasing prior to implementation.

Named entity recognition of pharmacokinetic parameters in the scientific literature.

The development of accurate predictions for a new drug's absorption, distribution, metabolism, and excretion profiles in the early stages of drug development is crucial due to high candidate failure rates. The absence of comprehensive, standardised, and updated pharmacokinetic (PK) repositories limits pre-clinical predictions and often requires searching through the scientific literature for PK parameter estimates from similar compounds. While text mining offers promising advancements in automatic PK parameter extraction, accurate Named Entity Recognition (NER) of PK terms remains a bottleneck due to limited resources. This work addresses this gap by introducing novel corpora and language models specifically designed for effective NER of PK parameters. Leveraging active learning approaches, we developed an annotated corpus containing over 4000 entity mentions found across the PK literature on PubMed. To identify the most effective model for PK NER, we fine-tuned and evaluated different NER architectures on our corpus. Fine-tuning BioBERT exhibited the best results, achieving a strict F 1 score of 90.37% in recognising PK parameter mentions, significantly outperforming heuristic approaches and models trained on existing corpora. To accelerate the development of end-to-end PK information extraction pipelines and improve pre-clinical PK predictions, the PK NER models and the labelled corpus were released open source at https://github.com/PKPDAI/PKNER .

PotatoG-DKB: a potato gene-disease knowledge base mined from biological literature.

Background: Potato is the fourth largest food crop in the world, but potato cultivation faces serious threats from various diseases and pests. Despite significant advancements in research on potato disease resistance, these findings are scattered across numerous publications. For researchers, obtaining relevant knowledge by reading and organizing a large body of literature is a time-consuming and labor-intensive process. Therefore, systematically extracting and organizing the relationships between potato genes and diseases from the literature to establish a potato gene-disease knowledge base is particularly important. Unfortunately, there is currently no such gene-disease knowledge base available. Methods: In this study, we constructed a Potato Gene-Disease Knowledge Base (PotatoG-DKB) using natural language processing techniques and large language models. We used PubMed as the data source and obtained 2,906 article abstracts related to potato biology, extracted entities and relationships between potato genes and related disease, and stored them in a Neo4j database. Using web technology, we also constructed the Potato Gene-Disease Knowledge Portal (PotatoG-DKP), an interactive visualization platform. Results: PotatoG-DKB encompasses 22 entity types (such as genes, diseases, species, etc.) of 5,206 nodes and 9,443 edges between entities (for example, gene-disease, pathogen-disease, etc.). PotatoG-DKP can intuitively display associative relationships extracted from literature and is a powerful assistant for potato biologists and breeders to understand potato pathogenesis and disease resistance. More details about PotatoG-DKP can be obtained at https://www.potatogd.com.cn/.

Use of SNOMED CT in Large Language Models: Scoping Review.

BACKGROUND: Large language models (LLMs) have substantially advanced natural language processing (NLP) capabilities but often struggle with knowledge-driven tasks in specialized domains such as biomedicine. Integrating biomedical knowledge sources such as SNOMED CT into LLMs may enhance their performance on biomedical tasks. However, the methodologies and effectiveness of incorporating SNOMED CT into LLMs have not been systematically reviewed. OBJECTIVE: This scoping review aims to examine how SNOMED CT is integrated into LLMs, focusing on (1) the types and components of LLMs being integrated with SNOMED CT, (2) which contents of SNOMED CT are being integrated, and (3) whether this integration improves LLM performance on NLP tasks. METHODS: Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, we searched ACM Digital Library, ACL Anthology, IEEE Xplore, PubMed, and Embase for relevant studies published from 2018 to 2023. Studies were included if they incorporated SNOMED CT into LLM pipelines for natural language understanding or generation tasks. Data on LLM types, SNOMED CT integration methods, end tasks, and performance metrics were extracted and synthesized. RESULTS: The review included 37 studies. Bidirectional Encoder Representations from Transformers and its biomedical variants were the most commonly used LLMs. Three main approaches for integrating SNOMED CT were identified: (1) incorporating SNOMED CT into LLM inputs (28/37, 76%), primarily using concept descriptions to expand training corpora; (2) integrating SNOMED CT into additional fusion modules (5/37, 14%); and (3) using SNOMED CT as an external knowledge retriever during inference (5/37, 14%). The most frequent end task was medical concept normalization (15/37, 41%), followed by entity extraction or typing and classification. While most studies (17/19, 89%) reported performance improvements after SNOMED CT integration, only a small fraction (19/37, 51%) provided direct comparisons. The reported gains varied widely across different metrics and tasks, ranging from 0.87% to 131.66%. However, some studies showed either no improvement or a decline in certain performance metrics. CONCLUSIONS: This review demonstrates diverse approaches for integrating SNOMED CT into LLMs, with a focus on using concept descriptions to enhance biomedical language understanding and generation. While the results suggest potential benefits of SNOMED CT integration, the lack of standardized evaluation methods and comprehensive performance reporting hinders definitive conclusions about its effectiveness. Future research should prioritize consistent reporting of performance comparisons and explore more sophisticated methods for incorporating SNOMED CT's relational structure into LLMs. In addition, the biomedical NLP community should develop standardized evaluation frameworks to better assess the impact of ontology integration on LLM performance.

Integration of big data analytics in the investigation of the relationship between acromegaly and cancer.

OBJECTIVE: To evaluate the association between acromegaly and cancer and different types of cancer by using natural language processing systems and big data analytics. MATERIAL AND METHODS: We conducted an observational, retrospective study utilizing data from the electronic health records (EHRs) of Hospital Universitario Puerta de Hierro Majadahonda, Madrid, Spain. Information from the EHRs was extracted using artificial intelligence techniques and analyzed using Savana Manager 4.0 software. RESULTS: Out of a total of 708,047 registered patients (54.7% females), 544 patients (0.08%; 330 women, 60.7%; mean age at diagnosis 53.0±15.8 yr) were diagnosed with acromegaly. The incidence of cancer was higher in patients with acromegaly vs those without this condition (7.7% vs 3.9%, p<0.001; OR, 2.047, 95%CI, 1.493-2.804). Male acromegalic patients had a higher prevalence of cancer vs females (57.1% vs 42.9%, p=0.012). A significantly higher prevalence of colorectal cancer (2.9% vs 1.4%, p=0.006), bladder cancer (1.1% vs 0.3%, p=0.005), and lymphoma (1.1% vs 0.3%, p=0.009) was observed in patients with acromegaly vs those without the condition. Acromegalic men had significantly higher prevalence rates of colorectal cancer (4.7% vs 1.3%, p=0.001), bladder cancer (2.8% vs 0.4%, p<0.001), breast cancer (0.9% vs 0.2%, p=0.042), gastric cancer (0.9% vs 0.1%, p=0.011), lymphoma (1.4% vs 0.3%, p=0.037), and liver cancer (0.9% vs 0.1%, p=0.012) vs non-acromegalic men. On the other hand, acromegalic women showed a higher prevalence of thyroid cancer (1.2% vs 0.4%, p=0.043) vs non-acromegalic women. CONCLUSION: Our study, based on artificial intelligence techniques and analysis of real-world data and information, revealed a significant association between acromegaly and cancer in our hospital population, mainly acromegalic men, with a higher frequency of colorectal cancer, bladder cancer and lymphoma in particular.

Estimating the prevalence of select non-communicable diseases in Saudi Arabia using a population-based sample: econometric analysis with natural language processing.

BACKGROUND: Non-communicable diseases (NCDs) are a major public health challenge globally, including in Saudi Arabia. However, measuring the true extent of NCD prevalence has been hampered by a paucity of nationally representative epidemiological studies. OBJECTIVES: Assess the prevalence of selected NCDs, using population-based electronic health records and applying novel analytical methods to identify cases of NCDs. DESIGN: Retrospective. SETTINGS: A large healthcare network in Saudi Arabia. PATIENTS AND METHODS: We included all beneficiaries aged 16 years or older (n=650 835[a]) and used the International Classification of Disease (ICD-10) codes, laboratory results, and associated medications to identify individuals with diabetes, obesity, hypertension, dyslipidemia, mental disorders, and injuries. For diabetes and hypertension, we used natural language processing (NLP) on clinical notes in the electronic health records. The prevalence of multimorbidity across age groups was also tabulated, and logistic regression was used to examine its association with glycemic control. MAIN OUTCOME MEASURES: The primary outcomes measured were the prevalence of diabetes, hypertension, and multimorbidity, and their association with glycemic control. SAMPLE SIZE: 650 835 individuals aged 16 years or older. RESULTS: The study population was relatively young, with 41.2% aged between 26 and 45 years, and around two-thirds were married. The prevalence of diabetes and hypertension was 18.5% (95% CI: 18.5-18.7) and 13.0% (95% CI: 12.9-13.1), respectively. Approximately 26.7% (95% CI: 26.7-26.8) of the population had multimorbidity, with levels increasing to 62.9% for those aged 65 or older. Multimorbidity was associated with a four-fold increase in the likelihood of poor glycemic control. NLP analysis suggested that the prevalence of diabetes or hypertension may be underestimated by no more than 1.5%. CONCLUSIONS: The study suggests a higher prevalence of NCDs than earlier national estimates. Electronic health records with regular analysis provide an opportunity to estimate changes in the prevalence of NCDs in Saudi Arabia. Health policies and interventions are needed to address the high levels of multimorbidity, which adversely impact glycemic control. LIMITATIONS: Retrospective design and reliance on electronic health records, which may not capture all cases of NCDs.

Engine of Innovation in Hospital Pharmacy: Applications and Reflections of ChatGPT.

Hospital pharmacy plays an important role in ensuring medical care quality and safety, especially in the area of drug information retrieval, therapy guidance, and drug-drug interaction management. ChatGPT is a powerful artificial intelligence language model that can generate natural-language texts. Here, we explored the applications and reflections of ChatGPT in hospital pharmacy, where it may enhance the quality and efficiency of pharmaceutical care. We also explored ChatGPT's prospects in hospital pharmacy and discussed its working principle, diverse applications, and practical cases in daily operations and scientific research. Meanwhile, the challenges and limitations of ChatGPT, such as data privacy, ethical issues, bias and discrimination, and human oversight, are discussed. ChatGPT is a promising tool for hospital pharmacy, but it requires careful evaluation and validation before it can be integrated into clinical practice. Some suggestions for future research and development of ChatGPT in hospital pharmacy are provided.

Practical Aspects of Using Large Language Models to Screen Abstracts for Cardiovascular Drug Development: Cross-Sectional Study.

Unlabelled: Cardiovascular drug development requires synthesizing relevant literature about indications, mechanisms, biomarkers, and outcomes. This short study investigates the performance, cost, and prompt engineering trade-offs of 3 large language models accelerating the literature screening process for cardiovascular drug development applications.

Validation of large language models for detecting pathologic complete response in breast cancer using population-based pathology reports.

AIMS: The primary goal of this study is to evaluate the capabilities of Large Language Models (LLMs) in understanding and processing complex medical documentation. We chose to focus on the identification of pathologic complete response (pCR) in narrative pathology reports. This approach aims to contribute to the advancement of comprehensive reporting, health research, and public health surveillance, thereby enhancing patient care and breast cancer management strategies. METHODS: The study utilized two analytical pipelines, developed with open-source LLMs within the healthcare system's computing environment. First, we extracted embeddings from pathology reports using 15 different transformer-based models and then employed logistic regression on these embeddings to classify the presence or absence of pCR. Secondly, we fine-tuned the Generative Pre-trained Transformer-2 (GPT-2) model by attaching a simple feed-forward neural network (FFNN) layer to improve the detection performance of pCR from pathology reports. RESULTS: In a cohort of 351 female breast cancer patients who underwent neoadjuvant chemotherapy (NAC) and subsequent surgery between 2010 and 2017 in Calgary, the optimized method displayed a sensitivity of 95.3% (95%CI: 84.0-100.0%), a positive predictive value of 90.9% (95%CI: 76.5-100.0%), and an F1 score of 93.0% (95%CI: 83.7-100.0%). The results, achieved through diverse LLM integration, surpassed traditional machine learning models, underscoring the potential of LLMs in clinical pathology information extraction. CONCLUSIONS: The study successfully demonstrates the efficacy of LLMs in interpreting and processing digital pathology data, particularly for determining pCR in breast cancer patients post-NAC. The superior performance of LLM-based pipelines over traditional models highlights their significant potential in extracting and analyzing key clinical data from narrative reports. While promising, these findings highlight the need for future external validation to confirm the reliability and broader applicability of these methods.

Machine learning and deep learning-based approach to categorize Bengali comments on social networks using fused dataset.

Through the advancement of the contemporary web and the rapid adoption of social media platforms such as YouTube, Twitter, and Facebook, for example, life has become much easier when dealing with certain highly personal problems. The far-reaching consequences of online harassment require immediate preventative steps to safeguard psychological wellness and scholarly achievement via detection at an earlier stage. This piece of writing aims to eliminate online harassment and create a criticism-free online environment. In the paper, we have used a variety of attributes to evaluate a large number of Bengali comments. We communicate cleansed data utilizing machine learning (ML) methods and natural language processing techniques, which must be followed using term frequency and reverse document frequency (TF-IDF) with a count vectorizer. In addition, we used tokenization with padding to feed our deep learning (DL) models. Using mathematical visualization and natural language processing, online bullying could be detected quickly. Multi-layer Perceptron (MLP), K-Nearest Neighbors (K-NN), Extreme Gradient Boosting (XGBoost), Adaptive Boosting Classifier (AdaBoost), Logistic Regression Classifier (LR), Random Forest Classifier (RF), Bagging Classifier, Stochastic Gradient Descent (SGD), Voting Classifier, and Stacking are employed in the research we conducted. We expanded our investigation to include different DL frameworks. Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Convolutional-Long Short-Term Memory (C-LSTM), and Bidirectional Long Short-Term Memory (BiLSTM) are all implemented. A large amount of data is required to precisely recognize harassing behavior. To rapidly recognize internet harassment written material, we combined two sets of data, producing 94,000 Bengali comments from different points of view. After understanding the ML and DL models, we can see that a hybrid model (MLP+SGD+LR) performed more effectively when compared to other models, its evaluation accuracy is 99.34%, precision is 99.34%, recall rate is 99.33%, and F1 score is 99.34% on multi-label class. For the binary classification model, we got 99.41% of accuracy.

A hybrid framework with large language models for rare disease phenotyping.

PURPOSE: Rare diseases pose significant challenges in diagnosis and treatment due to their low prevalence and heterogeneous clinical presentations. Unstructured clinical notes contain valuable information for identifying rare diseases, but manual curation is time-consuming and prone to subjectivity. This study aims to develop a hybrid approach combining dictionary-based natural language processing (NLP) tools with large language models (LLMs) to improve rare disease identification from unstructured clinical reports. METHODS: We propose a novel hybrid framework that integrates the Orphanet Rare Disease Ontology (ORDO) and the Unified Medical Language System (UMLS) to create a comprehensive rare disease vocabulary. SemEHR, a dictionary-based NLP tool, is employed to extract rare disease mentions from clinical notes. To refine the results and improve accuracy, we leverage various LLMs, including LLaMA3, Phi3-mini, and domain-specific models like OpenBioLLM and BioMistral. Different prompting strategies, such as zero-shot, few-shot, and knowledge-augmented generation, are explored to optimize the LLMs' performance. RESULTS: The proposed hybrid approach demonstrates superior performance compared to traditional NLP systems and standalone LLMs. LLaMA3 and Phi3-mini achieve the highest F1 scores in rare disease identification. Few-shot prompting with 1-3 examples yields the best results, while knowledge-augmented generation shows limited improvement. Notably, the approach uncovers a significant number of potential rare disease cases not documented in structured diagnostic records, highlighting its ability to identify previously unrecognized patients. CONCLUSION: The hybrid approach combining dictionary-based NLP tools with LLMs shows great promise for improving rare disease identification from unstructured clinical reports. By leveraging the strengths of both techniques, the method demonstrates superior performance and the potential to uncover hidden rare disease cases. Further research is needed to address limitations related to ontology mapping and overlapping case identification, and to integrate the approach into clinical practice for early diagnosis and improved patient outcomes.

Ascle-A Python Natural Language Processing Toolkit for Medical Text Generation: Development and Evaluation Study.

BACKGROUND: Medical texts present significant domain-specific challenges, and manually curating these texts is a time-consuming and labor-intensive process. To address this, natural language processing (NLP) algorithms have been developed to automate text processing. In the biomedical field, various toolkits for text processing exist, which have greatly improved the efficiency of handling unstructured text. However, these existing toolkits tend to emphasize different perspectives, and none of them offer generation capabilities, leaving a significant gap in the current offerings. OBJECTIVE: This study aims to describe the development and preliminary evaluation of Ascle. Ascle is tailored for biomedical researchers and clinical staff with an easy-to-use, all-in-one solution that requires minimal programming expertise. For the first time, Ascle provides 4 advanced and challenging generative functions: question-answering, text summarization, text simplification, and machine translation. In addition, Ascle integrates 12 essential NLP functions, along with query and search capabilities for clinical databases. METHODS: We fine-tuned 32 domain-specific language models and evaluated them thoroughly on 27 established benchmarks. In addition, for the question-answering task, we developed a retrieval-augmented generation (RAG) framework for large language models that incorporated a medical knowledge graph with ranking techniques to enhance the reliability of generated answers. Additionally, we conducted a physician validation to assess the quality of generated content beyond automated metrics. RESULTS: The fine-tuned models and RAG framework consistently enhanced text generation tasks. For example, the fine-tuned models improved the machine translation task by 20.27 in terms of BLEU score. In the question-answering task, the RAG framework raised the ROUGE-L score by 18% over the vanilla models. Physician validation of generated answers showed high scores for readability (4.95/5) and relevancy (4.43/5), with a lower score for accuracy (3.90/5) and completeness (3.31/5). CONCLUSIONS: This study introduces the development and evaluation of Ascle, a user-friendly NLP toolkit designed for medical text generation. All code is publicly available through the Ascle GitHub repository. All fine-tuned language models can be accessed through Hugging Face.

Disambiguating Clinical Abbreviations by One-to-All Classification: Algorithm Development and Validation Study.

Background: Electronic medical records store extensive patient data and serve as a comprehensive repository, including textual medical records like surgical and imaging reports. Their utility in clinical decision support systems is substantial, but the widespread use of ambiguous and unstandardized abbreviations in clinical documents poses challenges for natural language processing in clinical decision support systems. Efficient abbreviation disambiguation methods are needed for effective information extraction. Objective: This study aims to enhance the one-to-all (OTA) framework for clinical abbreviation expansion, which uses a single model to predict multiple abbreviation meanings. The objective is to improve OTA by developing context-candidate pairs and optimizing word embeddings in Bidirectional Encoder Representations From Transformers (BERT), evaluating the model's efficacy in expanding clinical abbreviations using real data. Methods: Three datasets were used: Medical Subject Headings Word Sense Disambiguation, University of Minnesota, and Chia-Yi Christian Hospital from Ditmanson Medical Foundation Chia-Yi Christian Hospital. Texts containing polysemous abbreviations were preprocessed and formatted for BERT. The study involved fine-tuning pretrained models, ClinicalBERT and BlueBERT, generating dataset pairs for training and testing based on Huang et al's method. Results: BlueBERT achieved macro- and microaccuracies of 95.41% and 95.16%, respectively, on the Medical Subject Headings Word Sense Disambiguation dataset. It improved macroaccuracy by 0.54%-1.53% compared to two baselines, long short-term memory and deepBioWSD with random embedding. On the University of Minnesota dataset, BlueBERT recorded macro- and microaccuracies of 98.40% and 98.22%, respectively. Against the baselines of Word2Vec + support vector machine and BioWordVec + support vector machine, BlueBERT demonstrated a macroaccuracy improvement of 2.61%-4.13%. Conclusions: This research preliminarily validated the effectiveness of the OTA method for abbreviation disambiguation in medical texts, demonstrating the potential to enhance both clinical staff efficiency and research effectiveness.

Building a Natural Language Interface for FHIR Clinical Terminology Server.

While Fast Healthcare Interoperability Resources (FHIR) clinical terminology server enables quick and easy search and retrieval of coded medical data, it still has some drawbacks. When searching, any typographical errors, variations in word forms, or deviations in word sequence might lead to incorrect search outcomes. For retrieval, queries to the server must strictly follow the FHIR application programming interface format, which requires users to know the syntax and remember the attribute codes they wish to retrieve. To improve its functionalities, a natural language interface was built, that harnesses the capabilities of two preeminent large language models, along with other cutting-edge technologies such as speech-to-text conversion, vector semantic searching, and conversational artificial intelligence. Preliminary evaluation shows promising results in building a natural language interface for the FHIR clinical terminology system.

The Normalization of Vaping on TikTok Using Computer Vision, Natural Language Processing, and Qualitative Thematic Analysis: Mixed Methods Study.

BACKGROUND: Social media posts that portray vaping in positive social contexts shape people's perceptions and serve to normalize vaping. Despite restrictions on depicting or promoting controlled substances, vape-related content is easily accessible on TikTok. There is a need to understand strategies used in promoting vaping on TikTok, especially among susceptible youth audiences. OBJECTIVE: This study seeks to comprehensively describe direct (ie, explicit promotional efforts) and indirect (ie, subtler strategies) themes promoting vaping on TikTok using a mixture of computational and qualitative thematic analyses of social media posts. In addition, we aim to describe how these themes might play a role in normalizing vaping behavior on TikTok for youth audiences, thereby informing public health communication and regulatory policies regarding vaping endorsements on TikTok. METHODS: We collected 14,002 unique TikTok posts using 50 vape-related hashtags (eg, #vapetok and #boxmod). Using the k-means unsupervised machine learning algorithm, we identified clusters and then categorized posts qualitatively based on themes. Next, we organized all videos from the posts thematically and extracted the visual features of each theme using 3 machine learning-based model architectures: residual network (ResNet) with 50 layers (ResNet50), Visual Geometry Group model with 16 layers, and vision transformer. We chose the best-performing model, ResNet50, to thoroughly analyze the image clustering output. To assess clustering accuracy, we examined 4.01% (441/10,990) of the samples from each video cluster. Finally, we randomly selected 50 videos (5% of the total videos) from each theme, which were qualitatively coded and compared with the machine-derived classification for validation. RESULTS: We successfully identified 5 major themes from the TikTok posts. Vape product marketing (1160/10,990, 8.28%) reflected direct marketing, while the other 4 themes reflected indirect marketing: TikTok influencer (3775/14,002, 26.96%), general vape (2741/14,002, 19.58%), vape brands (2042/14,002, 14.58%), and vaping cessation (1272/14,002, 9.08%). The ResNet50 model successfully classified clusters based on image features, achieving an average F1-score of 0.97, the highest among the 3 models. Qualitative content analyses indicated that vaping was depicted as a normal, routine part of daily life, with TikTok influencers subtly incorporating vaping into popular culture (eg, gaming, skateboarding, and tattooing) and social practices (eg, shopping sprees, driving, and grocery shopping). CONCLUSIONS: The results from both computational and qualitative analyses of text and visual data reveal that vaping is normalized on TikTok. Our identified themes underscore how everyday conversations, promotional content, and the influence of popular figures collectively contribute to depicting vaping as a normal and accepted aspect of daily life on TikTok. Our study provides valuable insights for regulatory policies and public health initiatives aimed at tackling the normalization of vaping on social media platforms.